Load packages
rm(list=ls(all=TRUE))
1source("_helpers/explorer.cat.r")
source("_helpers/explorer.num.r")
library(Boruta); library(caret); library(dplyr); library(MASS); library(partykit); library(pdp); library(randomForest); library(tree)Load the data
Add persistence measure: positive values show that the previous comparison was analytic and negative values show that it was synthetic
PERSISTENCE now only includes those cases within 1000 characters before the actual adjective, so adjust the PERSIST_FORM variable accordingly to only count those cases:
Add the Rhythm differences: if positive, the analytic pattern was better, if negative, the synthetic pattern was better
Add the Segment differences: if positive, the analytic patterns was better, if negative, the synthetic pattern was better
NO FILE COMPARISON ADJ_LEMMA
Min. :150557 Min. : 4703076 analytic :117 good : 80
1st Qu.:156950 1st Qu.: 4705601 synthetic:329 high : 36
Median :171688 Median : 4727066 bad : 21
Mean :189707 Mean :22743140 large : 19
3rd Qu.:219231 3rd Qu.:41786608 low : 17
Max. :234528 Max. :41855485 great : 16
(Other):257
CORPUS YEAR VARIETY NEWSPAPER WORD_COUNT
NOW2020 :217 Min. :2020 BrE:217 Daily Mail :108 Min. : 140
SAVE2020:229 1st Qu.:2020 LK :229 Daily Mirror:124 1st Qu.: 786
Median :2020 Daily News :105 Median :1194
Mean :2020 Independent :109 Mean :1561
3rd Qu.:2020 3rd Qu.:1803
Max. :2020 Max. :5629
FORM ADJ_LEN READABILITY LEXDIV
comparative:219 Min. : 3.000 Min. :-492.56 Min. :-203.241
superlative:227 1st Qu.: 4.000 1st Qu.:-210.52 1st Qu.: 2.881
Median : 4.000 Median : -19.68 Median : 17.506
Mean : 5.193 Mean : 256.85 Mean : 15.408
3rd Qu.: 6.000 3rd Qu.: 351.55 3rd Qu.: 34.824
Max. :13.000 Max. :4743.51 Max. : 61.193
STRESS_LAST_SYLL PERSIST_COMP PERSIST_FORM PERSIST_DIST SYNT_FUN
n:124 analytic : 84 comparative:160 Min. : 0.0 a :348
y:322 none :119 none :119 1st Qu.: 164.5 n : 26
synthetic:243 superlative:167 Median : 479.0 p : 70
Mean : 854.6 pn: 2
3rd Qu.:1160.0
Max. :8062.0
NA's :119
ADVMOD COMPL ZIPF_FREQ DPNOFREQ RHY_SCORE_POS
n:411 i: 11 Min. :2.438 Min. :0.0000938 Min. :0.00000
y: 35 n:398 1st Qu.:5.011 1st Qu.:0.0212153 1st Qu.:0.00000
p: 23 Median :5.524 Median :0.0323701 Median :0.00000
t: 14 Mean :5.380 Mean :0.0787586 Mean :0.07869
3rd Qu.:5.948 3rd Qu.:0.1563902 3rd Qu.:0.00000
Max. :6.377 Max. :0.2237505 Max. :1.00000
RHY_SCORE_ALT_POS SEG_SCORE SEG_SCORE_ALT FINAL_SEGMENT
Min. :0.0000 Min. :0.0000 Min. :0.0000 Min. :0.0000
1st Qu.:0.0000 1st Qu.:0.1769 1st Qu.:0.1824 1st Qu.:0.2500
Median :0.1667 Median :0.2378 Median :0.2500 Median :0.5000
Mean :0.1905 Mean :0.2407 Mean :0.2468 Mean :0.4105
3rd Qu.:0.3333 3rd Qu.:0.3122 3rd Qu.:0.3152 3rd Qu.:0.6250
Max. :1.0000 Max. :0.6250 Max. :0.5000 Max. :1.0000
PERSISTENCE PERSIST_FORM.1000 RHY_A RHY_S
Min. :-1.0000 comparative:114 Min. :0.0000 Min. :0.0000
1st Qu.:-0.6020 none :221 1st Qu.:0.0000 1st Qu.:0.0000
Median : 0.0000 superlative:111 Median :0.1250 Median :0.0000
Mean :-0.1638 Mean :0.1859 Mean :0.0833
3rd Qu.: 0.0000 3rd Qu.:0.3333 3rd Qu.:0.0000
Max. : 1.0000 Max. :1.0000 Max. :1.0000
RHY_DIFF SEG_A SEG_S SEG_DIFF
Min. :-1.0000 Min. :0.0000 Min. :0.0000 Min. :-0.17273
1st Qu.:-0.3333 1st Qu.:0.2000 1st Qu.:0.1667 1st Qu.:-0.05706
Median : 0.0000 Median :0.2547 Median :0.2301 Median :-0.03507
Mean :-0.1026 Mean :0.2520 Mean :0.2355 Mean :-0.01644
3rd Qu.: 0.0000 3rd Qu.:0.3206 3rd Qu.:0.3079 3rd Qu.: 0.01389
Max. : 0.6667 Max. :0.5000 Max. :0.6250 Max. : 0.32143
The variables initially included are: ADJ_LEN, ADVMOD, COMPARISON, COMPL, DPNOFREQ, FINAL_SEGMENT, FORM, LEXDIV, NEWSPAPER, PERSIST_FORM.1000, PERSISTENCE, READABILITY, RHY_DIFF, RHY_A, RHY_S, SEG_DIFF, SEG_A, SEG_S, STRESS_LAST_SYLL, SYNT_FUN, VARIETY, WORD_COUNT, ZIPF_FREQ
Look at COMPARISON: 26.2% analytic, 73.8% synthetic
Look at ADJ_LEN: 1st Qu = Median, but that doesn’t change in any of the suggested transformations, so leave it as is
$`Special data points`
Missing data Zeros Negatives
FALSE FALSE FALSE
$Summary
Min. 1st Qu. Median Mean 3rd Qu. Max.
3.000 4.000 4.000 5.193 6.000 13.000
$`Types and tokens`
Types Tokens
11 446
$`Freqs of freqs`
0 1 9 14 24 28 39 70 82 164
1 2 1 2 1 1 1 1 1 1
$`Smallest 'meaningful' difference`
[1] 1Look at ADVMOD: 92.2% no vs. 7.8% yes, very few adjectives are adverbially modified
$`Missing data`
[1] FALSE
$Frequencies
n y
411 35
$Percentages
n y
0.922 0.078
$`Freqs of freqs`
35 411
1 1
$`Types and tokens`
types (no NAs) Tokens (no NAs)
2 446 Look at COMPL: 2.3% to-infinitive, 84.7% no complement, 8% prepositional phrase, 5% than-phrase; few complements overall; conflate to to-infinitive and prepositional phrase vs. no complement and than-phrase (see Mondorf (2014) for why than-phrases are not complements).
$`Missing data`
[1] FALSE
$Frequencies
i n p t
11 398 23 14
$Percentages
i n p t
0.025 0.892 0.052 0.031
$`Freqs of freqs`
11 14 23 398
1 1 1 1
$`Types and tokens`
types (no NAs) Tokens (no NAs)
4 446 node), split, n, deviance, yval, (yprob)
* denotes terminal node
1) root 446 513.30 synthetic ( 0.2623 0.7377 )
2) x$COMPL: i,p 34 46.66 synthetic ( 0.4412 0.5588 ) *
3) x$COMPL: n,t 412 461.20 synthetic ( 0.2476 0.7524 ) *Look at COMPL.confl: 7.6% yes vs. 92.4% no
$`Missing data`
[1] FALSE
$Frequencies
y n
34 412
$Percentages
y n
0.076 0.924
$`Freqs of freqs`
34 412
1 1
$`Types and tokens`
types (no NAs) Tokens (no NAs)
2 446 Look at DPNOFREQ: looks okay
$`Special data points`
Missing data Zeros Negatives
FALSE FALSE FALSE
$Summary
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.0000938 0.0212153 0.0323701 0.0787586 0.1563902 0.2237505
$`Types and tokens`
Types Tokens
149 446
$`Freqs of freqs`
0 1 2 3 4 5 7 9 10 11 12 18 26 38 41
1 84 32 7 4 6 4 3 3 1 1 1 1 1 1
$`Smallest 'meaningful' difference`
[1] 5.49e-06Look at FINAL_SEGMENT: fine
$`Special data points`
Missing data Zeros Negatives
"FALSE" "TRUE:100" "FALSE"
$Summary
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.0000 0.2500 0.5000 0.4105 0.6250 1.0000
$`Types and tokens`
Types Tokens
14 446
$`Freqs of freqs`
0 2 11 12 13 17 19 20 43 80 100 102
1 2 1 2 2 1 1 1 1 1 1 1
$`Smallest 'meaningful' difference`
[1] 0.04166667Look at FORM: 49.1% comparative vs. 50.9% superlative
$`Missing data`
[1] FALSE
$Frequencies
comparative superlative
219 227
$Percentages
comparative superlative
0.491 0.509
$`Freqs of freqs`
219 227
1 1
$`Types and tokens`
types (no NAs) Tokens (no NAs)
2 446 Look at LEXDIV: one pretty low value, winsorize to lowest value in boxplot that is not an outlier (-43.82754)
$`Special data points`
Missing data Zeros Negatives
"FALSE" "FALSE" "TRUE:86"
$Summary
Min. 1st Qu. Median Mean 3rd Qu. Max.
-203.241 2.881 17.506 15.408 34.824 61.193
$`Types and tokens`
Types Tokens
120 446
$`Freqs of freqs`
0 1 2 3 4 5 6 7 8 9 10 11 12 14 16 25
1 43 19 12 12 9 5 7 2 3 1 1 2 2 1 1
$`Smallest 'meaningful' difference`
[1] 0.0196487
[1] -43.82754Look at LEXDIV.win: better
$`Special data points`
Missing data Zeros Negatives
"FALSE" "FALSE" "TRUE:86"
$Summary
Min. 1st Qu. Median Mean 3rd Qu. Max.
-43.828 2.881 17.506 16.629 34.824 61.193
$`Types and tokens`
Types Tokens
113 446
$`Freqs of freqs`
0 1 2 3 4 5 6 7 8 9 10 11 12 14 16 22 25
1 37 19 12 11 9 5 7 2 3 1 1 1 2 1 1 1
$`Smallest 'meaningful' difference`
[1] 0.0196487Look at NEWSPAPER: 24.2% Daily Mail vs. 27.8% Daily Mirror vs. 23.5% Daily News vs. 24.4% Independent
$`Missing data`
[1] FALSE
$Frequencies
Daily Mail Daily Mirror Daily News Independent
108 124 105 109
$Percentages
Daily Mail Daily Mirror Daily News Independent
0.242 0.278 0.235 0.244
$`Freqs of freqs`
105 108 109 124
1 1 1 1
$`Types and tokens`
types (no NAs) Tokens (no NAs)
4 446 Look at PERSIST_FORM.1000: 25.6% comparative vs. 49.6% none vs. 24.9% superlative
$`Missing data`
[1] FALSE
$Frequencies
comparative none superlative
114 221 111
$Percentages
comparative none superlative
0.256 0.496 0.249
$`Freqs of freqs`
111 114 221
1 1 1
$`Types and tokens`
types (no NAs) Tokens (no NAs)
3 446 Look at PERSISTENCE: about half of the values are 0, I could change that by raising the max. limit of PERSIST_DIST (right now it’s 1000 characters), but there are still 26.8% of data points without Persistence so this wouldn’t be fixed entirely; more negative than positive values because there are more synthetic comparisons
$`Special data points`
Missing data Zeros Negatives
"FALSE" "TRUE:221" "TRUE:163"
$Summary
Min. 1st Qu. Median Mean 3rd Qu. Max.
-1.0000 -0.6020 0.0000 -0.1638 0.0000 1.0000
$`Types and tokens`
Types Tokens
189 446
$`Freqs of freqs`
0 1 2 3 8 221
1 161 22 4 1 1
$`Smallest 'meaningful' difference`
[1] 0.001Look at READABILITY: log it
$`Special data points`
Missing data Zeros Negatives
"FALSE" "FALSE" "TRUE:236"
$Summary
Min. 1st Qu. Median Mean 3rd Qu. Max.
-492.56 -210.52 -19.68 256.85 351.55 4743.51
$`Types and tokens`
Types Tokens
120 446
$`Freqs of freqs`
0 1 2 3 4 5 6 7 8 9 10 11 12 14 16 25
1 43 19 12 12 9 5 7 2 3 1 1 2 2 1 1
$`Smallest 'meaningful' difference`
[1] 0.2674029Look at READABILITY.log: better
$`Special data points`
Missing data Zeros Negatives
"FALSE" "FALSE" "TRUE:229"
$Summary
Min. 1st Qu. Median Mean 3rd Qu. Max.
-8.9442 -7.7178 -4.2990 -0.1076 8.4576 12.2117
$`Types and tokens`
Types Tokens
120 446
$`Freqs of freqs`
0 1 2 3 4 5 6 7 8 9 10 11 12 14 16 25
1 43 19 12 12 9 5 7 2 3 1 1 2 2 1 1
$`Smallest 'meaningful' difference`
[1] 0.001060231Look at RHY_DIFF: a lot of zeros but none of the suggested transformations would fix that
$`Special data points`
Missing data Zeros Negatives
"FALSE" "TRUE:222" "TRUE:182"
$Summary
Min. 1st Qu. Median Mean 3rd Qu. Max.
-1.0000 -0.3333 0.0000 -0.1026 0.0000 0.6667
$`Types and tokens`
Types Tokens
51 446
$`Freqs of freqs`
0 1 2 3 4 5 6 7 11 13 17 37 66 222
1 23 13 3 1 1 1 1 1 1 1 1 1 1
$`Smallest 'meaningful' difference`
[1] 0.00297619Look at RHY_A: looks fine
$`Special data points`
Missing data Zeros Negatives
"FALSE" "TRUE:207" "FALSE"
$Summary
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.0000 0.0000 0.1250 0.1859 0.3333 1.0000
$`Types and tokens`
Types Tokens
26 446
$`Freqs of freqs`
0 1 2 3 4 5 6 20 24 53 90 207
1 7 7 2 2 1 2 1 1 1 1 1
$`Smallest 'meaningful' difference`
[1] 0.00297619Look at RHY_S: again a lot of zeros
$`Special data points`
Missing data Zeros Negatives
"FALSE" "TRUE:336" "FALSE"
$Summary
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.0000 0.0000 0.0000 0.0833 0.0000 1.0000
$`Types and tokens`
Types Tokens
24 446
$`Freqs of freqs`
0 1 2 3 5 6 14 25 33 336
1 10 7 1 1 1 1 1 1 1
$`Smallest 'meaningful' difference`
[1] 0.005952381Look at SEG_DIFF: fine
$`Special data points`
Missing data Zeros Negatives
"FALSE" "TRUE:37" "TRUE:277"
$Summary
Min. 1st Qu. Median Mean 3rd Qu. Max.
-0.17273 -0.05706 -0.03507 -0.01644 0.01389 0.32143
$`Types and tokens`
Types Tokens
264 446
$`Freqs of freqs`
0 1 2 3 4 5 6 7 37
1 175 41 22 7 4 4 2 1
$`Smallest 'meaningful' difference`
[1] 4.768e-06Look at SEG_A: fine
$`Special data points`
Missing data Zeros Negatives
"FALSE" "TRUE:10" "FALSE"
$Summary
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.0000 0.2000 0.2547 0.2520 0.3206 0.5000
$`Types and tokens`
Types Tokens
172 446
$`Freqs of freqs`
0 1 2 3 4 5 6 7 8 9 10 11 13 16 24
1 95 27 17 10 5 1 4 4 2 2 1 2 1 1
$`Smallest 'meaningful' difference`
[1] 2.3527e-05Look at SEG_S: fine
$`Special data points`
Missing data Zeros Negatives
"FALSE" "TRUE:19" "FALSE"
$Summary
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.0000 0.1667 0.2301 0.2355 0.3079 0.6250
$`Types and tokens`
Types Tokens
175 446
$`Freqs of freqs`
0 1 2 3 4 5 6 7 8 10 11 12 16 19
1 103 30 12 6 6 3 3 1 2 4 1 2 2
$`Smallest 'meaningful' difference`
[1] 1.8038e-05Look at STRESS_LAST_SYLL: 27.8% no vs. 72.2% yes
$`Missing data`
[1] FALSE
$Frequencies
n y
124 322
$Percentages
n y
0.278 0.722
$`Freqs of freqs`
124 322
1 1
$`Types and tokens`
types (no NAs) Tokens (no NAs)
2 446 Look at SYNT_FUN: 78% attributive vs. 5.8% nominal vs. 15.7% predicative vs. 0.4% post-nominal; conflate to predicative and post-nominal vs. attributive and nominal
$`Missing data`
[1] FALSE
$Frequencies
a n p pn
348 26 70 2
$Percentages
a n p pn
0.780 0.058 0.157 0.004
$`Freqs of freqs`
2 26 70 348
1 1 1 1
$`Types and tokens`
types (no NAs) Tokens (no NAs)
4 446 node), split, n, deviance, yval, (yprob)
* denotes terminal node
1) root 446 513.30 synthetic ( 0.2623 0.7377 )
2) x$SYNT_FUN: p,pn 72 99.59 synthetic ( 0.4722 0.5278 ) *
3) x$SYNT_FUN: a,n 374 395.90 synthetic ( 0.2219 0.7781 ) *Look at SYNT_FUN.confl: 83.9% attributive vs. 16.1% predicative
$`Missing data`
[1] FALSE
$Frequencies
a p
374 72
$Percentages
a p
0.839 0.161
$`Freqs of freqs`
72 374
1 1
$`Types and tokens`
types (no NAs) Tokens (no NAs)
2 446 Look at VARIETY: 48.7% BrE vs. 51.3% LK
$`Missing data`
[1] FALSE
$Frequencies
BrE LK
217 229
$Percentages
BrE LK
0.487 0.513
$`Freqs of freqs`
217 229
1 1
$`Types and tokens`
types (no NAs) Tokens (no NAs)
2 446 Look at WORD_COUNT: log
$`Special data points`
Missing data Zeros Negatives
FALSE FALSE FALSE
$Summary
Min. 1st Qu. Median Mean 3rd Qu. Max.
140 786 1194 1561 1803 5629
$`Types and tokens`
Types Tokens
115 446
$`Freqs of freqs`
0 1 2 3 4 5 6 7 8 9 10 11 12 14 16 25
1 40 19 11 13 7 4 6 2 3 1 2 2 3 1 1
$`Smallest 'meaningful' difference`
[1] 1Look at WORD_COUNT.log: better
$`Special data points`
Missing data Zeros Negatives
FALSE FALSE FALSE
$Summary
Min. 1st Qu. Median Mean 3rd Qu. Max.
7.129 9.618 10.222 10.213 10.816 12.459
$`Types and tokens`
Types Tokens
115 446
$`Freqs of freqs`
0 1 2 3 4 5 6 7 8 9 10 11 12 14 16 25
1 40 19 11 13 7 4 6 2 3 1 2 2 3 1 1
$`Smallest 'meaningful' difference`
[1] 0.00135528Look at ZIPF_FREQ: skewed to the left… boxcox
$`Special data points`
Missing data Zeros Negatives
FALSE FALSE FALSE
$Summary
Min. 1st Qu. Median Mean 3rd Qu. Max.
2.438 5.011 5.524 5.380 5.948 6.377
$`Types and tokens`
Types Tokens
146 446
$`Freqs of freqs`
0 1 2 3 4 5 7 9 10 11 12 18 26 38 41
1 81 30 8 5 6 4 3 3 1 1 1 1 1 1
$`Smallest 'meaningful' difference`
[1] 5.1168e-05
[1] 2Look at ZIPF_FREQ.trans:
$`Special data points`
Missing data Zeros Negatives
FALSE FALSE FALSE
$Summary
Min. 1st Qu. Median Mean 3rd Qu. Max.
2.473 12.056 14.756 14.282 17.192 19.836
$`Types and tokens`
Types Tokens
146 446
$`Freqs of freqs`
0 1 2 3 4 5 7 9 10 11 12 18 26 38 41
1 81 30 8 5 6 4 3 3 1 1 1 1 1 1
$`Smallest 'meaningful' difference`
[1] 0.000319287The categorical variables
x$VARIETYxADVMOD <- x$VARIETY:x$ADVMOD
x$VARIETYxCOMPL.confl <- x$VARIETY:x$COMPL.confl
x$VARIETYxFORM <- x$VARIETY:x$FORM
x$VARIETYxNEWSPAPER <- x$VARIETY:x$NEWSPAPER
x$VARIETYxPERSIST_FORM.1000 <- x$VARIETY:x$PERSIST_FORM.1000
x$VARIETYxSTRESS_LAST_SYLL <- x$VARIETY:x$STRESS_LAST_SYLL
x$VARIETYxSYNT_FUN.confl <- x$VARIETY:x$SYNT_FUN.conflThe numeric variables - I went with ctree() because for RHY_DIFF the tree() function made too many splits to compute them & e.g. the 15 splits for ZIPF_FREQ are just not interpretable
ADJ_LEN:#(t1 <- tree(COMPARISON ~ ADJ_LEN, data=x)) # 4 levels
plot(t1 <- ctree(COMPARISON ~ ADJ_LEN, data=x)) # 4 levels
Split ADJ_LEN into 4 levels:
DPNOFREQ#(t2 <- tree(COMPARISON ~ DPNOFREQ, data=x)) # 15 levels
plot(t2 <- ctree(COMPARISON ~ DPNOFREQ, data=x)) # 3 levels
Split DPNOFREQ into 3 levels:
FINAL_SEGMENT:#(t3 <- tree(COMPARISON ~ FINAL_SEGMENT, data=x)) # 5 levels
plot(t3 <- ctree(COMPARISON ~ FINAL_SEGMENT, data=x))
Split FINAL_SEGMENT into 5 levels:
LEXDIV.win:# (t3 <- tree(COMPARISON ~ LEXDIV.win, data=x)) # 7 levels
plot(t3 <- ctree(COMPARISON ~ LEXDIV.win, data=x))
No splits for LEXDIV.win
PERSISTENCE:#(t4 <- tree(COMPARISON ~ PERSISTENCE, data=x)) # 2 levels
plot(t4 <- ctree(COMPARISON ~ PERSISTENCE, data=x)) # 2 levels
Split PERSISTENCE into 2 levels:
READABILITY.log:#(t5 <- tree(COMPARISON ~ READABILITY, data=x)) # 3 levels
plot(t5 <- ctree(COMPARISON ~ READABILITY.log, data=x))
No split for READABILITY.log
RHY_DIFF:#(t6 <- tree(COMPARISON ~ RHY_DIFF, data=x))
plot(t6 <- ctree(COMPARISON ~ RHY_DIFF, data=x)) # 4 levels
Using ctree(), split RHY_DIFF into 4 levels:
RHY_A:#(t7 <- tree(COMPARISON ~ RHY_A, data=x)) # 7 levels
plot(t7 <- ctree(COMPARISON ~ RHY_A, data=x)) # 4 levels
Split RHY_A into 4 levels:
RHY_S:Split RHY_S into 2 levels:
SEG_DIFFnode), split, n, deviance, yval, (yprob)
* denotes terminal node
1) root 446 513.30 synthetic ( 0.2623 0.7377 )
2) SEG_DIFF < 0.154006 433 505.30 synthetic ( 0.2702 0.7298 )
4) SEG_DIFF < 0.0586081 398 448.70 synthetic ( 0.2513 0.7487 )
8) SEG_DIFF < 0.0414377 385 441.00 synthetic ( 0.2597 0.7403 ) *
9) SEG_DIFF > 0.0414377 13 0.00 synthetic ( 0.0000 1.0000 ) *
5) SEG_DIFF > 0.0586081 35 48.49 synthetic ( 0.4857 0.5143 )
10) SEG_DIFF < 0.0663055 8 0.00 analytic ( 1.0000 0.0000 ) *
11) SEG_DIFF > 0.0663055 27 34.37 synthetic ( 0.3333 0.6667 ) *
3) SEG_DIFF > 0.154006 13 0.00 synthetic ( 0.0000 1.0000 ) *
No split for SEG_DIFF.
SEG_Anode), split, n, deviance, yval, (yprob)
* denotes terminal node
1) root 446 513.300 synthetic ( 0.26233 0.73767 )
2) SEG_A < 0.364358 399 481.000 synthetic ( 0.29073 0.70927 ) *
3) SEG_A > 0.364358 47 9.679 synthetic ( 0.02128 0.97872 ) *
No split for SEG_A.
SEG_S#(t11 <- tree(COMPARISON ~ SEG_S, data=x)) # 9 levels
plot(t11 <- ctree(COMPARISON ~ SEG_S, data=x))
Split SEG_S into 2 levels
WORD_COUNT.log:#(t12 <- tree(COMPARISON ~ WORD_COUNT, data=x)) # 8 levels
plot(t12 <- ctree(COMPARISON ~ WORD_COUNT.log, data=x))
No splits for WORD_COUNT.log
ZIPF_FREQ.trans:#(t13 <- tree(COMPARISON ~ ZIPF_FREQ, data=x)) # 15 levels
plot(t13 <- ctree(COMPARISON ~ ZIPF_FREQ.trans, data=x))
Split ZIPF_FREQ into 4 levels:
x$VARIETYxADJ_LEN.cat <- x$VARIETY:x$ADJ_LEN.cat
x$VARIETYxDPNOFREQ.cat <- x$VARIETY:x$DPNOFREQ.cat
x$VARIETYxFINAL_SEGMENT.cat <- x$VARIETY:x$FINAL_SEGMENT.cat
x$VARIETYxPERSISTENCE.cat <- x$VARIETY:x$PERSISTENCE.cat
x$VARIETYxRHY_DIFF.cat <- x$VARIETY:x$RHY_DIFF.cat
x$VARIETYxRHY_A.cat <- x$VARIETY:x$RHY_A.cat
x$VARIETYxRHY_S.cat <- x$VARIETY:x$RHY_S.cat
x$VARIETYxSEG_S.cat <- x$VARIETY:x$SEG_S.cat
x$VARIETYxZIPF_FREQ.cat <- x$VARIETY:x$ZIPF_FREQ.catVariabes to include in the model: ADJ_LEN, DPNOFREQ, FINAL_SEGMENT, LEXDIV.win, PERSISTENCE, READABILITY.log, RHY_DIFF, RHY_A, RHY_S, SEG_DIFF, SEG_A, SEG_A, WORD_COUNT.log, ZIPF_FREQ.trans, ADVMOD, COMPL.confl, FORM, NEWSPAPER, PERSIST_FORM.1000, STRESS_LAST_SYLL, SYNT_FUN.confl, VARIETY, VARIETYxADJ_LEN.cat, VARIETYxDPNOFREQ.cat , VARIETYxFINAL_SEGMENT.cat, VARIETYxPERSISTENCE.cat, VARIETYxRHY_DIFF.cat, VARIETYxRHY_A.cat, VARIETYxRHY_S.cat, VARIETYxSEG_S.cat, VARIETYxZIPF_FREQ.cat, VARIETYxADVMOD,VARIETYxCOMPL.confl, VARIETYxFORM,VARIETYxNEWSPAPER, VARIETYxPERSIST_FORM.1000, VARIETYxSTRESS_LAST_SYLL, VARIETYxSYNT_FUN.confl
Compute the two baselines: 73.8% is the one to beat
c("baseline 1"=baseline.1 <- max(prop.table(table(x$COMPARISON))),
"baseline 2"=baseline.2 <- sum(prop.table(table(x$COMPARISON))^2))baseline 1 baseline 2
0.7376682 0.6129723 Variable Selection Boruta suggests:
ADJ_LEN, DPNOFREQ, FINAL_SEGMENT, LEXDIV.win, PERSISTENCE, READABILITY.log, RHY_DIFF, RHY_A, SEG_DIFF, SEG_A, SEG_A, WORD_COUNT.log, ZIPF_FREQ.trans, STRESS_LAST_SYLL, SYNT_FUN.confl, VARIETYxADJ_LEN.cat, VARIETYxDPNOFREQ.cat , VARIETYxFINAL_SEGMENT.cat, VARIETYxRHY_DIFF.cat, VARIETYxRHY_A.cat, VARIETYxRHY_S.cat, VARIETYxSEG_S.cat, VARIETYxZIPF_FREQ.cat, VARIETYxADVMOD,VARIETYxCOMPL.confl, VARIETYxFORM, VARIETYxPERSIST_FORM.1000, VARIETYxSTRESS_LAST_SYLL, VARIETYxSYNT_FUN.confl
Exclude:
ADVMOD, COMPL.confl, FORM, NEWSPAPER, PERSIST_FORM.1000, RHY_S, VARIETY, VARIETYxPERSISTENCE.cat, VARIETYxNEWSPAPER, VARIETYxFORM
set.seed(sum(utf8ToInt("All the Young Dudes")))
predictors <- Boruta(COMPARISON ~ ADJ_LEN + ADVMOD + COMPL.confl + DPNOFREQ + FINAL_SEGMENT + FORM + LEXDIV.win + NEWSPAPER + PERSIST_FORM.1000 + PERSISTENCE + READABILITY.log + RHY_DIFF + RHY_A + RHY_S + SEG_DIFF + SEG_A + SEG_S + STRESS_LAST_SYLL + SYNT_FUN.confl + VARIETY + WORD_COUNT.log + ZIPF_FREQ.trans + VARIETYxADJ_LEN.cat + VARIETYxDPNOFREQ.cat + VARIETYxFINAL_SEGMENT.cat + VARIETYxPERSISTENCE.cat + VARIETYxRHY_DIFF.cat + VARIETYxRHY_A.cat + VARIETYxRHY_S.cat + VARIETYxSEG_S.cat + VARIETYxZIPF_FREQ.cat + VARIETYxADVMOD + VARIETYxCOMPL.confl + VARIETYxFORM + VARIETYxNEWSPAPER + VARIETYxPERSIST_FORM.1000 + VARIETYxSTRESS_LAST_SYLL + VARIETYxSYNT_FUN.confl, data=x, maxRuns=200)
attStats(predictors) meanImp medianImp minImp maxImp
VARIETYxSYNT_FUN.confl 3.54578446 3.4638071 1.64966908 5.6261129
VARIETYxSTRESS_LAST_SYLL 12.44683773 12.4118911 11.20183095 14.3581703
VARIETYxPERSIST_FORM.1000 2.53623705 2.5048933 1.16104112 4.5512077
VARIETYxNEWSPAPER 1.93250701 1.7795970 0.88231118 3.3473140
VARIETYxFORM 2.35650839 2.3187569 0.53085762 3.8383047
VARIETYxCOMPL.confl 2.40712788 2.4046307 1.11279383 3.9329763
VARIETYxADVMOD 2.31672896 2.3179899 0.49107400 4.0874927
VARIETYxZIPF_FREQ.cat 8.07115224 8.0270622 6.61081010 9.3860573
VARIETYxSEG_S.cat 5.26019305 5.2166157 3.13493937 7.3416357
VARIETYxRHY_S.cat 2.60748819 2.5915440 1.22571884 4.1086403
VARIETYxRHY_A.cat 6.15898981 6.2087533 4.51892337 7.6294173
VARIETYxRHY_DIFF.cat 4.92380164 4.9571398 3.39509718 6.4740550
VARIETYxPERSISTENCE.cat 2.22943727 2.2681966 0.43409144 3.3567433
VARIETYxFINAL_SEGMENT.cat 9.26702857 9.2726826 7.66759675 11.2296263
VARIETYxDPNOFREQ.cat 6.98797017 6.9885349 5.68062101 8.5891499
VARIETYxADJ_LEN.cat 15.76010524 15.7633681 13.89847136 17.4338296
ZIPF_FREQ.trans 13.41959471 13.4160578 11.86299506 15.1310351
WORD_COUNT.log 4.28080533 4.3368132 2.19988328 6.4535804
VARIETY 1.89480225 1.9730096 1.00496698 3.1065494
SYNT_FUN.confl 2.63116152 2.7921912 0.10110122 5.0710036
STRESS_LAST_SYLL 17.21235294 17.1881314 15.31306868 19.0004399
SEG_S 6.72334641 6.7311847 4.37666027 8.9601071
SEG_A 5.99421647 6.0866438 2.93829494 7.7660516
SEG_DIFF 5.34021651 5.3714519 3.05801860 7.9290785
RHY_S 1.02660591 0.8626261 -0.22760646 2.6753267
RHY_A 7.35328277 7.3631153 5.84394696 8.7590434
RHY_DIFF 5.69787288 5.7327761 3.44364789 7.0657322
READABILITY.log 5.15818113 5.1819261 3.43467867 6.6844643
PERSISTENCE 2.98263567 3.0849548 -0.05460303 5.5023577
PERSIST_FORM.1000 0.56706758 0.7603471 -0.77446357 1.3490702
NEWSPAPER 1.17849512 1.3634910 -1.05759155 2.6664365
LEXDIV.win 6.65022033 6.6921415 3.50813903 9.4436205
FORM 0.70913984 0.8126717 -1.45905062 2.6178364
FINAL_SEGMENT 13.24988218 13.2527389 11.85797181 14.7474660
DPNOFREQ 10.37066890 10.3572321 8.84269509 12.3910781
COMPL.confl 0.67879077 0.6255756 -1.04513401 2.1400496
ADVMOD 0.03530495 0.1743451 -1.29804535 0.9400113
ADJ_LEN 23.24500180 23.1763745 21.63927155 25.2229926
normHits decision
VARIETYxSYNT_FUN.confl 0.87939698 Confirmed
VARIETYxSTRESS_LAST_SYLL 1.00000000 Confirmed
VARIETYxPERSIST_FORM.1000 0.51758794 Tentative
VARIETYxNEWSPAPER 0.02512563 Rejected
VARIETYxFORM 0.45226131 Tentative
VARIETYxCOMPL.confl 0.44221106 Tentative
VARIETYxADVMOD 0.44221106 Tentative
VARIETYxZIPF_FREQ.cat 1.00000000 Confirmed
VARIETYxSEG_S.cat 0.98994975 Confirmed
VARIETYxRHY_S.cat 0.54271357 Tentative
VARIETYxRHY_A.cat 1.00000000 Confirmed
VARIETYxRHY_DIFF.cat 0.98492462 Confirmed
VARIETYxPERSISTENCE.cat 0.38190955 Tentative
VARIETYxFINAL_SEGMENT.cat 1.00000000 Confirmed
VARIETYxDPNOFREQ.cat 1.00000000 Confirmed
VARIETYxADJ_LEN.cat 1.00000000 Confirmed
ZIPF_FREQ.trans 1.00000000 Confirmed
WORD_COUNT.log 0.95477387 Confirmed
VARIETY 0.06030151 Rejected
SYNT_FUN.confl 0.56281407 Tentative
STRESS_LAST_SYLL 1.00000000 Confirmed
SEG_S 1.00000000 Confirmed
SEG_A 0.98994975 Confirmed
SEG_DIFF 0.98994975 Confirmed
RHY_S 0.01005025 Rejected
RHY_A 1.00000000 Confirmed
RHY_DIFF 1.00000000 Confirmed
READABILITY.log 0.98994975 Confirmed
PERSISTENCE 0.66331658 Confirmed
PERSIST_FORM.1000 0.00000000 Rejected
NEWSPAPER 0.01507538 Rejected
LEXDIV.win 1.00000000 Confirmed
FORM 0.00000000 Rejected
FINAL_SEGMENT 1.00000000 Confirmed
DPNOFREQ 1.00000000 Confirmed
COMPL.confl 0.00000000 Rejected
ADVMOD 0.00000000 Rejected
ADJ_LEN 1.00000000 Confirmedcollector <- matrix(rep(NA, 60), ncol=10, dimnames=list(
NTREE=ntree.vals <- c(500, 1000, 1500, 2000, 2500, 3000),
MTRY=mtry.vals <- 1:10))
for(k in seq(ntree.vals)){
for(j in seq(mtry.vals)){
seedy <- set.seed(sum(utf8ToInt("All the Young Dudes")))
collector[k,j] <- randomForest(
COMPARISON ~
ADJ_LEN + DPNOFREQ + FINAL_SEGMENT + LEXDIV.win + PERSISTENCE + READABILITY.log +
RHY_DIFF + RHY_A + SEG_DIFF + SEG_A + SEG_S + STRESS_LAST_SYLL + SYNT_FUN.confl +
WORD_COUNT.log + ZIPF_FREQ.trans + VARIETYxADJ_LEN.cat + VARIETYxDPNOFREQ.cat +
VARIETYxFINAL_SEGMENT.cat + VARIETYxRHY_DIFF.cat + VARIETYxRHY_A.cat + VARIETYxRHY_S.cat +
VARIETYxSEG_S.cat + VARIETYxZIPF_FREQ.cat + VARIETYxADVMOD + VARIETYxCOMPL.confl +
VARIETYxPERSIST_FORM.1000 + VARIETYxSTRESS_LAST_SYLL + VARIETYxSYNT_FUN.confl,
data=x,
ntree=ntree.vals[k], mtry=mtry.vals[j],
importance=TRUE)$err.rate[ntree.vals[k], "OOB"]
#cat(seedy, ntree.vals[k], mtry.vals[j], collector[k,j], "\n", sep="\t")
}
}
mtry <- which(t(collector)==min(t(collector)), arr.ind=TRUE)[1,1]
ntree <- which(t(collector)==min(t(collector)), arr.ind=TRUE)[1,2]*500 smallest collector for mtry=5 and ntree=500
set.seed(sum(utf8ToInt("All the Young Dudes")))
(rf.1 <- randomForest(COMPARISON ~
ADJ_LEN + DPNOFREQ + FINAL_SEGMENT + LEXDIV.win + PERSISTENCE + READABILITY.log +
RHY_DIFF + RHY_A + SEG_DIFF + SEG_A + SEG_S + STRESS_LAST_SYLL + SYNT_FUN.confl +
WORD_COUNT.log + ZIPF_FREQ.trans + VARIETYxADJ_LEN.cat + VARIETYxDPNOFREQ.cat +
VARIETYxFINAL_SEGMENT.cat + VARIETYxRHY_DIFF.cat + VARIETYxRHY_A.cat + VARIETYxRHY_S.cat +
VARIETYxSEG_S.cat + VARIETYxZIPF_FREQ.cat + VARIETYxADVMOD + VARIETYxCOMPL.confl +
VARIETYxPERSIST_FORM.1000 + VARIETYxSTRESS_LAST_SYLL + VARIETYxSYNT_FUN.confl,
data=x,
ntree=500,
mtry=5,
importance=TRUE))
Call:
randomForest(formula = COMPARISON ~ ADJ_LEN + DPNOFREQ + FINAL_SEGMENT + LEXDIV.win + PERSISTENCE + READABILITY.log + RHY_DIFF + RHY_A + SEG_DIFF + SEG_A + SEG_S + STRESS_LAST_SYLL + SYNT_FUN.confl + WORD_COUNT.log + ZIPF_FREQ.trans + VARIETYxADJ_LEN.cat + VARIETYxDPNOFREQ.cat + VARIETYxFINAL_SEGMENT.cat + VARIETYxRHY_DIFF.cat + VARIETYxRHY_A.cat + VARIETYxRHY_S.cat + VARIETYxSEG_S.cat + VARIETYxZIPF_FREQ.cat + VARIETYxADVMOD + VARIETYxCOMPL.confl + VARIETYxPERSIST_FORM.1000 + VARIETYxSTRESS_LAST_SYLL + VARIETYxSYNT_FUN.confl, data = x, ntree = 500, mtry = 5, importance = TRUE)
Type of random forest: classification
Number of trees: 500
No. of variables tried at each split: 5
OOB estimate of error rate: 1.79%
Confusion matrix:
analytic synthetic class.error
analytic 113 4 0.03418803
synthetic 4 325 0.01215805(c.m <- table(OBS=x$COMPARISON, PREDS=x$PREDS.CAT.rf1)); c( # confusion matrix & its eval
"Class. acc." =mean(x$COMPARISON==x$PREDS.CAT.rf1, na.rm=TRUE),
"Prec. for synthetic"=c.m["synthetic","synthetic"] / sum(c.m[ ,"synthetic"]),
"Rec. for synthetic" =c.m["synthetic","synthetic"] / sum(c.m["synthetic",]),
"Prec. for analytic"=c.m["analytic","analytic"] / sum(c.m[ ,"analytic"]),
"Rec. for analytic" =c.m["analytic","analytic"] / sum(c.m["analytic",])) PREDS
OBS analytic synthetic
analytic 113 4
synthetic 4 325 Class. acc. Prec. for synthetic Rec. for synthetic Prec. for analytic
0.9820628 0.9878419 0.9878419 0.9658120
Rec. for analytic
0.9658120 MeanDecreaseAccuracy MeanDecreaseGini
ADJ_LEN 1.034095e-01 37.1440855
DPNOFREQ 7.585457e-03 3.1235906
FINAL_SEGMENT 1.276964e-02 8.4235328
LEXDIV.win 3.715063e-03 3.0412576
PERSISTENCE 1.380123e-03 1.8226580
READABILITY.log 2.489495e-03 1.8775345
RHY_DIFF 1.846570e-03 0.9541832
RHY_A 2.602820e-03 1.3885271
SEG_DIFF 2.796271e-03 2.9088197
SEG_A 1.957821e-03 1.7146716
SEG_S 3.559805e-03 2.6198574
STRESS_LAST_SYLL 4.358566e-02 20.6168262
SYNT_FUN.confl 5.259399e-04 0.2920680
WORD_COUNT.log 1.652742e-03 2.0106882
ZIPF_FREQ.trans 1.679121e-02 6.5149179
VARIETYxADJ_LEN.cat 7.738788e-02 29.1257391
VARIETYxDPNOFREQ.cat 4.855285e-03 1.9980827
VARIETYxFINAL_SEGMENT.cat 1.505212e-02 12.2140002
VARIETYxRHY_DIFF.cat 1.702025e-03 1.0668374
VARIETYxRHY_A.cat 3.789232e-03 2.1697736
VARIETYxRHY_S.cat 2.998041e-04 0.3283484
VARIETYxSEG_S.cat 3.149952e-03 1.1055828
VARIETYxZIPF_FREQ.cat 1.531515e-02 9.2415441
VARIETYxADVMOD 1.792046e-05 0.2107844
VARIETYxCOMPL.confl 2.563686e-04 0.2171749
VARIETYxPERSIST_FORM.1000 1.436808e-03 1.1977105
VARIETYxSTRESS_LAST_SYLL 3.847270e-02 18.0130404
VARIETYxSYNT_FUN.confl 1.366823e-03 0.6454828dotchart(sort(varimps[,1]), pch=4, xlab='Mean Decrease in Accuracy', main='Variable Importance Plot')
Confusion Matrix and Statistics
Reference
Prediction analytic synthetic
analytic 113 4
synthetic 4 325
Accuracy : 0.9821
95% CI : (0.965, 0.9922)
No Information Rate : 0.7377
P-Value [Acc > NIR] : <2e-16
Kappa : 0.9537
Mcnemar's Test P-Value : 1
Sensitivity : 0.9658
Specificity : 0.9878
Pos Pred Value : 0.9658
Neg Pred Value : 0.9878
Prevalence : 0.2623
Detection Rate : 0.2534
Detection Prevalence : 0.2623
Balanced Accuracy : 0.9768
'Positive' Class : analytic
Save Variable importance scores
#png("03a_VarIMPPlot.png", width=40, height=30, units="cm", res=300)
dotchart(sort(varimps[,1]), pch=4, xlab='Mean Decrease in Accuracy', main='Variable Importance Plot')
Plot for ADJ_LEN:
(pd.cas <- partial( # make pd.c contain partial dependence scores
object=rf.1, # from this forest
pred.var="ADJ_LEN", # for this predictor
which.class=2, # for the 2nd level of the response
train=x,
prob=TRUE)) ADJ_LEN yhat
1 3 0.8152601
2 4 0.8152601
3 5 0.8100628
4 6 0.6037713
5 7 0.5324574
6 8 0.4989013
7 9 0.4963901
8 10 0.4961839
9 11 0.4961839
10 12 0.4961839
11 13 0.4961839tab.cas <- prop.table(table(x$ADJ_LEN))
#png("03b_pd-adjlen.png", widt=18, height=13, units="cm", res=300)
plot(
main="Partial dep. of COMPARISON on ADJ_LEN",
type="b", pch=16,
xlab="Adjective Length in Characters",
ylab=substitute(paste('Probability of ', italic('synthetic'), ' comparison')),
ylim=c(0,1),
x=pd.cas$ADJ_LEN,
y=pd.cas$yhat,
cex=1+tab.cas*10
)
#abline(h=partial(object=rf.1, pred.var = 'VARIETY', which.class=2, train=x, prob=TRUE)[1,2], lty=2)
abline(h=sum(x$COMPARISON=='synthetic')/nrow(x), lty=2)
lines(lowess(pd.cas$yhat ~ pd.cas$ADJ_LEN), lwd=6, col="#BCBCBCB0")
Plot for VARIETYxADJ_LEN.cat:
(pd.corpustrigger <- partial(object=rf.1,
pred.var='VARIETYxADJ_LEN.cat',
which.class=2, train=x,
prob=TRUE)) VARIETYxADJ_LEN.cat yhat
1 BrE:(0,4] 0.7914305
2 BrE:(4,5] 0.7831570
3 BrE:(5,7] 0.6180807
4 BrE:(7,13] 0.5765336
5 LK:(0,4] 0.7983812
6 LK:(4,5] 0.7775516
7 LK:(5,7] 0.6371570
8 LK:(7,13] 0.5821973b <- matrix(data=pd.corpustrigger$yhat, nrow=4)
rownames(b) <- levels(x$ADJ_LEN.cat)
colnames(b) <- levels(x$VARIETY)
#png("03c_pd-varietyxadjlen.png", widt=18, height=13, units="cm", res=300)
plot(x=0, ylim=c(0.5,1), xlim=c(0,3), xaxt='n', bty='n', pch='',
xlab='Adjective length in characters',
ylab=substitute(paste('Probability of ', italic('synthetic'), ' comparison')),
main='Partial dep. of COMPARISON on VARIETYxADJ_LEN.cat',
cex.main=1.5);grid()
axis(1, at=0:3, labels=levels(x$ADJ_LEN.cat))
#abline(h=partial(object=rf.1, pred.var='VARIETY', which.class=2, train=x, prob=TRUE)[1,2], lty=2)
abline(h=sum(x$COMPARISON=='synthetic')/nrow(x), lty=2)
points(x=0:3, y=b[,2], pch=16, col=alpha('#004F86', alpha=0.7), cex=3)
points(x=0:3, y=b[,1], pch=16, col=alpha('#7BC8FF', alpha=0.7), cex=3)
legend('top', legend=c('BrE', 'SLE'), fill=c('#7BC8FF', '#004F86'), ncol=2, xjust=0.5, yjust=0.5)
Plot for STRESS_LAST_SYLL:
(pd.corpustrigger <- partial(object=rf.1,
pred.var='STRESS_LAST_SYLL',
which.class=2, train=x,
prob=TRUE)) STRESS_LAST_SYLL yhat
1 n 0.6305247
2 y 0.7606323b <- matrix(data=pd.corpustrigger$yhat, nrow=2)
rownames(b) <- levels(x$STRESS_LAST_SYLL)
#png("03d_pd-stresslastsyll.png", width=22.5, height=15, units="cm", res=300)
plot(x=0, ylim=c(0.5,1), xlim=c(0,1), xaxt='n', bty='n', pch='',
xlab='Stress on the last syllable of the adjective lemma',
ylab=substitute(paste('Probability of ', italic('synthetic'), 'comparison')),
main='Partial dep. of COMPARISON on STRESS_LAST_SYLL',
cex.main=1.5);grid()
axis(1, at=0:1, labels=c('no', 'yes'))
#abline(h=partial(object=rf.1, pred.var='VARIETY', which.class=2, train=x, prob=TRUE)[1,2], lty=2)
abline(h=sum(x$COMPARISON=='synthetic')/nrow(x), lty=2)
points(x=0:1, y=b[,1], pch=16, col=alpha('#004F86', alpha=0.7), cex=3)
Plot for VARIETYxSTRESS_LAST_SYLL:
(pd.corpustrigger <- partial(object=rf.1,
pred.var='VARIETYxSTRESS_LAST_SYLL',
which.class=2, train=x,
prob=TRUE)) VARIETYxSTRESS_LAST_SYLL yhat
1 BrE:n 0.6576906
2 BrE:y 0.7624126
3 LK:n 0.6456278
4 LK:y 0.7597489b <- matrix(data=pd.corpustrigger$yhat, nrow=2)
rownames(b) <- levels(x$STRESS_LAST_SYLL)
colnames(b) <- levels(x$VARIETY)
#png("03e_pd-varietyxstresslastsyll.png", widt=18, height=13, units="cm", res=300)
plot(x=0, ylim=c(0.5,1), xlim=c(0,1), xaxt='n', bty='n', pch='',
xlab='Stress on the last syllable of the adjective lemma',
ylab=substitute(paste('Probability of ', italic('synthetic'), ' comparison')),
main='Partial dep. of COMPARISON on VARIETYxSTRESS_LAST_SYLL',
cex.main=1);grid()
axis(1, at=0:1, labels=c('no', 'yes'))
#abline(h=partial(object=rf.1, pred.var='VARIETY', which.class=2, train=x, prob=TRUE)[1,2], lty=2)
abline(h=sum(x$COMPARISON=='synthetic')/nrow(x), lty=2)
points(x=0:1, y=b[,2], pch=16, col=alpha('#004F86', alpha=0.7), cex=3)
points(x=0:1, y=b[,1], pch=16, col=alpha('#7BC8FF', alpha=0.7), cex=3)
legend('top', legend=c('BrE', 'SLE'), fill=c('#7BC8FF', '#004F86'), ncol=2, xjust=0.5, yjust=0.5)
Plot for VAREITYxZIPF_FREQ.cat:
(pd.corpustrigger <- partial(object=rf.1,
pred.var='VARIETYxZIPF_FREQ.cat',
which.class=2, train=x,
prob=TRUE)) VARIETYxZIPF_FREQ.cat yhat
1 BrE:[-0.5,13.293] 0.7229193
2 BrE:(13.293,14.756] 0.7448251
3 BrE:(14.756,14.942] 0.7315830
4 BrE:(14.942,24] 0.7465830
5 LK:[-0.5,13.293] 0.6843632
6 LK:(13.293,14.756] 0.7469372
7 LK:(14.756,14.942] 0.7244439
8 LK:(14.942,24] 0.7456099b <- matrix(data=pd.corpustrigger$yhat, nrow=4)
rownames(b) <- levels(x$ZIPF_FREQ.cat)
colnames(b) <- levels(x$VARIETY)
#png("03f_pd-varietyxfrequency.png", widt=18, height=13, units="cm", res=300)
plot(x=0, ylim=c(0.5,1), xlim=c(-0.2,3.2), xaxt='n', bty='n', pch='',
xlab='Frequency of the Adjective Lemma',
ylab=substitute(paste('Probability of ', italic('synthetic'), ' comparison')),
main='Partial dep. of COMPARISON on VARIETYxZIPF_FREQ.cat',
cex.main=1);grid()
axis(1, at=0:3, labels=c("[0,13.293]", levels(x$ZIPF_FREQ.cat)[2:4]))
#abline(h=partial(object=rf.1, pred.var='VARIETY', which.class=2, train=x, prob=TRUE)[1,2], lty=2)
abline(h=sum(x$COMPARISON=='synthetic')/nrow(x), lty=2)
points(x=c(0.05, 1.05, 2.05, 3.05), y=b[,2], pch=16, col=alpha('#004F86', alpha=0.7), cex=3)
points(x=c(-0.05, 0.95, 1.95, 2.95), y=b[,1], pch=16, col=alpha('#7BC8FF', alpha=0.7), cex=3)
legend('top', legend=c('BrE', 'SLE'), fill=c('#7BC8FF', '#004F86'), ncol=2, xjust=0.5, yjust=0.5)
Plot for VARIETYxFINAL_SEGMENT.cat:
(pd.corpustrigger <- partial(object=rf.1,
pred.var='VARIETYxFINAL_SEGMENT.cat',
which.class=2, train=x,
prob=TRUE)) VARIETYxFINAL_SEGMENT.cat yhat
1 BrE:0 0.7511256
2 BrE:(0,0.333] 0.7414215
3 BrE:(0.333,0.625] 0.7367668
4 BrE:(0.625,0.667] 0.6889058
5 BrE:(0.667,1] 0.7008117
6 LK:0 0.7480493
7 LK:(0,0.333] 0.7569148
8 LK:(0.333,0.625] 0.7343363
9 LK:(0.625,0.667] 0.7314350
10 LK:(0.667,1] 0.6906637b <- matrix(data=pd.corpustrigger$yhat, nrow=5)
rownames(b) <- levels(x$FINAL_SEGMENT.cat)
colnames(b) <- levels(x$VARIETY)
#png("03g_pd-varietyxfinalsegment.png", widt=18, height=13, units="cm", res=300)
plot(x=0, ylim=c(0.5,1), xlim=c(-0.2,4.2), xaxt='n', bty='n', pch='',
xlab='Similarity of the Final Segment to the Synthetic Ending',
ylab=substitute(paste('Probability of ', italic('synthetic'), ' comparison')),
main='Partial dep. of COMPARISON on VARIETYxFINAL_SEGMENT.cat',
cex.main=1);grid()
axis(1, at=0:4, labels=levels(x$FINAL_SEGMENT.cat))
#abline(h=partial(object=rf.1, pred.var='VARIETY', which.class=2, train=x, prob=TRUE)[1,2], lty=2)
abline(h=sum(x$COMPARISON=='synthetic')/nrow(x), lty=2)
points(x=c(0.05,1.05,2.05,3.05,4.05), y=b[,2], pch=16, col=alpha('#004F86', alpha=0.7), cex=3)
points(x=c(-0.05,0.95,1.95,2.95,3.95), y=b[,1], pch=16, col=alpha('#7BC8FF', alpha=0.7), cex=3)
legend('top', legend=c('BrE', 'SLE'), fill=c('#7BC8FF', '#004F86'), ncol=2, xjust=0.5, yjust=0.5)
R version 4.3.0 (2023-04-21 ucrt)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 11 x64 (build 26100)
Matrix products: default
locale:
[1] LC_COLLATE=German_Germany.utf8 LC_CTYPE=German_Germany.utf8
[3] LC_MONETARY=German_Germany.utf8 LC_NUMERIC=C
[5] LC_TIME=German_Germany.utf8
time zone: Europe/Berlin
tzcode source: internal
attached base packages:
[1] grid stats graphics grDevices utils datasets methods
[8] base
other attached packages:
[1] car_3.1-2 carData_3.0-5 tree_1.0-43
[4] randomForest_4.7-1.1 pdp_0.8.1 partykit_1.2-20
[7] mvtnorm_1.2-2 libcoin_1.0-9 MASS_7.3-58.4
[10] dplyr_1.1.2 caret_6.0-94 lattice_0.21-8
[13] ggplot2_3.5.0 Boruta_8.0.0
loaded via a namespace (and not attached):
[1] tidyselect_1.2.0 timeDate_4022.108 farver_2.1.1
[4] fastmap_1.1.1 pROC_1.18.4 digest_0.6.31
[7] rpart_4.1.19 timechange_0.2.0 lifecycle_1.0.3
[10] survival_3.5-5 magrittr_2.0.3 compiler_4.3.0
[13] rlang_1.1.1 tools_4.3.0 utf8_1.2.3
[16] yaml_2.3.7 data.table_1.14.8 knitr_1.50
[19] htmlwidgets_1.6.2 plyr_1.8.8 abind_1.4-5
[22] withr_2.5.0 purrr_1.0.1 nnet_7.3-18
[25] stats4_4.3.0 fansi_1.0.4 e1071_1.7-16
[28] colorspace_2.1-0 future_1.33.0 globals_0.16.2
[31] scales_1.3.0 iterators_1.0.14 cli_3.6.1
[34] inum_1.0-5 rmarkdown_2.22 generics_0.1.3
[37] rstudioapi_0.14 future.apply_1.11.0 reshape2_1.4.4
[40] proxy_0.4-27 stringr_1.5.0 splines_4.3.0
[43] parallel_4.3.0 vctrs_0.6.3 hardhat_1.3.0
[46] Matrix_1.5-4 jsonlite_1.8.5 Formula_1.2-5
[49] listenv_0.9.0 foreach_1.5.2 gower_1.0.1
[52] recipes_1.0.6 glue_1.6.2 parallelly_1.36.0
[55] codetools_0.2-19 lubridate_1.9.2 stringi_1.7.12
[58] gtable_0.3.3 munsell_0.5.0 tibble_3.2.1
[61] pillar_1.9.0 htmltools_0.5.5 ipred_0.9-14
[64] lava_1.7.2.1 R6_2.5.1 evaluate_0.21
[67] class_7.3-21 Rcpp_1.0.10 nlme_3.1-162
[70] prodlim_2023.03.31 ranger_0.16.0 xfun_0.51
[73] ModelMetrics_1.2.2.2 pkgconfig_2.0.3